Google DeepMind has introduced Veo 2, an advanced AI model designed to generate high-quality video content. As the successor to the original Veo, Veo 2 offers significant improvements, including the ability to create video clips longer than two minutes and at resolutions up to 4K (4096 x 2160 pixels). This sets it apart from OpenAI’s Sora, which is limited to shorter videos and lower resolutions.
Currently, Veo 2 is accessible through Google’s experimental VideoFX tool, but video outputs are restricted to 720p resolution and eight-second durations. Google plans to expand access to VideoFX later this week, providing more users with the chance to explore the AI’s capabilities.
Veo 2 generates videos from text prompts or a combination of text and reference images. The updated model enhances the realism of motion, fluid dynamics, light properties, and human expressions, offering more cinematic effects and nuanced details. However, Eli Collins, Vice President of Product at DeepMind, acknowledged that there is still room for improvement, particularly in areas such as coherence, complex movements, and achieving higher realism in generated content. Collins emphasized that the team will continue refining Veo 2 based on user feedback over the coming months.
Related topics:
Google Expands Vertex AI with Veo, a Revolutionary Video Generation Model
Alessandra Alessio Joins StackAdapt as APAC Marketing Director
Brussels Investigates Google and Meta Over Secret Ad Deal Targeting Teens